Test shapiro-wilk

[guil]

Bonjour,

Je souhaite faire un test shapiro-wilk mais mes données sont sup à 5000. Auriez vous une astuce svp?

Merci par avance
Cordialement

[nirgrahamuk]

method : https://www.tandfonline.com/doi/abs/10.1080/02664769723828
R implementation : Index of /src/contrib/Archive/PoweR

[Andrzej]

Is this test any different to that one:
https://rdrr.io/r/stats/shapiro.test.html

Arguments
x 	

a numeric vector of data values. Missing values are allowed, but the number of non-missing values must be between 3 and 5000.

I can't install it (PoweR) in R 4.3.

[technocrat]

@guil

shapiro-wilk

Stackoverflow a une bonne discussion sur Shapiro-Wilk et d'autres tests de normalité

library(nortest)
# Anderson Darling
ad.test(runif(10000, min = 2, max = 4))
#> 
#>  Anderson-Darling normality test
#> 
#> data:  runif(10000, min = 2, max = 4)
#> A = 112.94, p-value < 2.2e-16
# Cramer-von Mises
cvm.test(runif(10000, min = 2, max = 4))
#> Warning in cvm.test(runif(10000, min = 2, max = 4)): p-value is smaller than
#> 7.37e-10, cannot be computed more accurately
#> 
#>  Cramer-von Mises normality test
#> 
#> data:  runif(10000, min = 2, max = 4)
#> W = 15.472, p-value = 7.37e-10
# Lilliefors (Kolmogorov-Smirnov)
lillie.test(runif(10000, min = 2, max = 4))
#> 
#>  Lilliefors (Kolmogorov-Smirnov) normality test
#> 
#> data:  runif(10000, min = 2, max = 4)
#> D = 0.059554, p-value < 2.2e-16
# Pearson chi-square test for normality
pearson.test(runif(10000, min = 2, max = 4))
#> 
#>  Pearson chi-square normality test
#> 
#> data:  runif(10000, min = 2, max = 4)
#> P = 2769.5, p-value < 2.2e-16

^{Created on 2023-06-28 with reprex v2.0.2}